Air-sea gas transfer velocity for oxygen derived from float data

We estimated the air-sea gas transfer velocity for oxygen using three consecutive years (Sept. 2003 to Aug. 2006) of high-quality oxygen measurements from profiling floats in the central Labrador Sea. Mixed layer oxygen concentrations exhibit strong seasonality characterized by biologically and ther...

Full description

Bibliographic Details
Published in:Journal of Geophysical Research
Main Authors: Kihm, Christoph, Körtzinger, Arne
Format: Article in Journal/Newspaper
Language:English
Published: AGU (American Geophysical Union) 2010
Subjects:
Online Access:https://oceanrep.geomar.de/id/eprint/10176/
https://oceanrep.geomar.de/id/eprint/10176/1/2009JC006077-pip.pdf
https://oceanrep.geomar.de/id/eprint/10176/2/2009JC006077.pdf
https://doi.org/10.1029/2009JC006077
Description
Summary:We estimated the air-sea gas transfer velocity for oxygen using three consecutive years (Sept. 2003 to Aug. 2006) of high-quality oxygen measurements from profiling floats in the central Labrador Sea. Mixed layer oxygen concentrations exhibit strong seasonality characterized by biologically and thermally driven evasion during spring/summer and invasion during fall/winter caused by cooling and ventilation of oxygen-deficient subsurface waters. Mixed layer oxygen budgets entirely excluding the spring bloom period are employed to estimate the air-sea transfer velocity for oxygen. By using co-located wind speed data acquired by scatterometry from the QuikSCAT satellite, wind speed dependent parameterizations for the air-sea gas transfer velocity k660 (CO2 at 20◦C and salinity 35) are established and compared with prominent parameterizations from the literature. Quadratic, cubic and quartic functions are fitted to the data for short-term and long-term wind speed averages separately. In both cases the quadratic functions yield the poorest fit to the observations. Overall, the stronger curvature of the cubic functions provides the best fit, while the quartic function also fits the data less well. Our results generally confirm the stronger wind speed dependencies among the suite of published parameterizations. Also the better fits found for cubic function points at the strong importance of very high wind speed for airsea gas exchange of O2.